def _find_client_id(self):
res = get_req(url='https://soundcloud.com/', headers=self._headers)
for src in reversed(re.findall(r'<script[^>]+src="([^"]+)"',
res.text)):
res_js_script = get_req(url=src, headers=self._headers)
if res_js_script:
cliend_id = re.findall(r'client_id\s*:\s*"([0-9a-zA-Z]{32})"',
res_js_script.text)
if cliend_id:
return cliend_id[0]
raise SoundCloudException('Unable to extract client id')
def _extract_info_dict_for_sets(self, *args, **kwargs):
query = kwargs.get('query', {}).copy()
query['client_id'] = kwargs.get('client_id')
url_info = kwargs.get('url_info')
response = get_req(url=url_info, headers=self._headers, params=query)
res_json = response.json()
_title = res_json.get('title')
_id = res_json.get('id')
_tracks = res_json.get('tracks')
sys.stdout.write(fg + '[' + fc + '*' + fg +
'] : Playlist %s found %s track.\n' %
(_title, len(_tracks)))
if _tracks:
for ele, track in enumerate(_tracks):
_id_track = track.get('id')
dict_info_track = self._extract_info_dict(
query=query,
url_info=self._API_V2_BASE + 'resolve?url=' +
self._API_BASE + 'tracks/' + str(_id_track),
client_id=self._cliend_id,
ele=ele + 1)
self._download_track(dict_info_track=dict_info_track,
_show_all_info=self._show_all_info,
ele=ele + 1)
def _extract_playlist(self, url, query):
query['client_id'] = self._cliend_id
query.update({
'limit': 2000000000,
'linked_partitioning': '1',
})
res = get_req(url=url, headers=self._headers, params=query)
if not isinstance(res.json(), dict):
return
return res.json()
def _get_info_user(self, url, query):
query['client_id'] = self._cliend_id
res = get_req(url=url, headers=self._headers, params=query)
if not isinstance(res.json(), dict):
return
return res.json()
def _extract_url_transcodings(self, url_trans, query):
res = get_req(url=url_trans, params=query, headers=self._headers)
return res.json()
def _extract_info_dict(self, *args, **kwargs):
ele = kwargs.get('ele') or ''
text = fg + '\r[' + fc + '*' + fg + '] : Extracting info of track %s ... ' % (
ele)
spinner(text=text)
query = kwargs.get('query', {}).copy()
query['client_id'] = kwargs.get('client_id')
url_info = kwargs.get('url_info')
response = get_req(url=url_info, headers=self._headers, params=query)
res_json = response.json()
format_urls = set()
formats = []
_description = res_json.get('description')
_title = res_json.get('title')
_artwork_url = res_json.get('artwork_url')
_id = res_json.get('id')
formats.append({
'title': _title,
'id': _id,
'artwork_url': _artwork_url,
'description': _description,
})
def invalid_url(url):
return not url or url in format_urls or re.search(
r'/(?:preview|playlist)/0/30/', url)
def add_format(f, protocol):
mobj = re.search(r'\.(?P<abr>\d+)\.(?P<ext>[0-9a-z]{3,4})(?=[/?])',
stream_url)
if mobj:
for k, v in mobj.groupdict().items():
if not f.get(k):
f[k] = v
format_id_list = []
if protocol:
format_id_list.append(protocol)
for k in ('ext', 'abr'):
v = f.get(k)
if v:
format_id_list.append(v)
abr = f.get('abr')
if abr:
f['abr'] = int(abr)
f.update({
'format_id':
'_'.join(format_id_list),
'protocol':
'm3u8_native' if protocol == 'hls' else 'http',
})
formats.append(f)
transcodings = try_get(res_json, lambda x: x['media']['transcodings'],
list) or []
for trans in transcodings:
if not isinstance(trans, dict):
continue
format_url = trans.get('url')
stream = self._extract_url_transcodings(format_url, query)
if not isinstance(stream, dict):
continue
stream_url = stream.get('url')
spinner(text=text)
if invalid_url(stream_url):
continue
format_urls.add(stream_url)
stream_format = trans.get('format') or {}
protocol = stream_format.get('protocol')
if protocol != 'hls' and '/hls' in format_url:
protocol = 'hls'
ext = None
preset = trans.get('preset')
if preset:
ext = preset.split('_')[0]
if ext not in KNOWN_EXTENSIONS:
ext = mimetype2ext(stream_format.get('mime_type'))
add_format({
'url': stream_url,
'ext': ext,
}, 'http' if protocol == 'progressive' else protocol)
sys.stdout.write(fg + '\r[' + fc + '*' + fg +
'] : Extracting info of track %s. (done)\n' % (ele))
return formats